Despite reported pain prevalences as high as 75 - 80'0, studies consistently report that pain is under-diagnosed and under-treated in the nursing home. Demented individuals, who constitute the majority of nursing home residents, are particularly at risk for under-diagnosis and under-treatment of pain because of their decreased self-report capacity. Currently, no pain assessment instrument exists that is simultaneously brief, easy to administer, reliable, well-validated, and inclusive of a variety of pain behaviors typical of dementia patients. The goal of this project is to comprehensively evaluate and refine the NOPAIN (Non- communicative patients Observational Pain Assessment INstrument), a brief standardized pain assessment instrument to be used with severely demented individuals. The NOPAIN items are currently being developed with the assistance of an expert clinical, research, and psychometric panel in a small funded study already underway that will be completed April, 2001. A completed pilot study has demonstrated the feasibility of our assessment approach, partnering a brief standardized physical examination with observation of specific categories of pain behaviors. The Specific Aims of the proposed project are: (1) to demonstrate the clinical feasibility of the NOPAIN, including ease of administration, time required for administration, and ability of untrained raters to administer the instrument; (2) to demonstrate the construct validity of the NOPAIN in severely demented individuals; (3) to demonstrate the inter-racer, internal consistency, and test-retest reliability of the NOPAIN in severely demented individuals. Participants will be moderately to severely demented individuals from 5 - 7 nursing homes. NOPAIN validity evidence will be obtained by comparing NOPAIN ratings of a low pain probability group (no pain- related conditions or current analgesic use) to a high pain probability group (a pain-related condition and current analgesic use) (n_ = 150 for both groups). Validity evidence will also be obtained by comparing the change between NOPAIN ratings taken one hour apart in a control group to the change between pre- and post-medication administration NOPAIN ratings in a group receiving a prn analgesic medication (n_ = 35 for both groups). Inter-later reliability will be evaluated by both comparing four raters who will observe NOPAIN administrations live, and by comparing 20 raters who will observe videotaped administrations.